He I Ultraviolet Photoelectron Spectroscopy of Benzene and ... · in Supersonic Molecular Beams Using Photoelectron Imaging ... pulsed molecular beam depending on the experiment

Published: March 17, 2011

r 2011 American Chemical Society 2953 dx.doi.org/10.1021/jp1098574 | J. Phys. Chem. A 2011, 115, 2953–2965

ARTICLE

pubs.acs.org/JPCA

He I Ultraviolet Photoelectron Spectroscopy of Benzene and Pyridinein Supersonic Molecular Beams Using Photoelectron ImagingSuet-Yi Liu,†,‡ Koutayba Alnama,† Jun Matsumoto,†,^ Kiyoshi Nishizawa,† Hiroshi Kohguchi,†,§

Yuan-Pern Lee,‡ and Toshinori Suzuki*,†,‡,||

†Chemical Dynamics Laboratory, RIKEN Advanced Science Institute, 2-1 Hirosawa, Wako 351-0198, Japan‡Department of Applied Chemistry, National Chiao Tung University, Hsinchu 30010, Taiwan§Department of Chemistry, School of Science, Hiroshima University, 1-3-1 Kagamiyama, Higashi-Hiroshima 739-8526, Japan

)Department of Chemistry, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan

bS Supporting Information

ABSTRACT: We performed He I ultraviolet photoelectronspectroscopy (UPS) of jet-cooled aromatic molecules using anewly developed photoelectron imaging (PEI) spectrometer.The PEI spectrometer can measure photoelectron spectraand photoelectron angular distributions at a considerablyhigher efficiency than a conventional spectrometer that uses ahemispherical energy analyzer. One technical problem withPEI is its relatively high susceptibility to background elec-trons generated by scattered He I radiation. To reduce thisproblem, we designed a new electrostatic lens that interceptsbackground photoelectrons emitted from the repeller plate toward the imaging detector. An energy resolution (ΔE/E) of 0.735% atE = 5.461 eV is demonstrated with He I radiation. The energy resolution is limited by the size of the ionization region. Trajectorycalculations indicate that the system is capable of achieving an energy resolution of 0.04% with a laser if the imaging resolution is notlimited. Experimental results are presented for jet-cooled benzene and pyridine, and they are compared with results in the literature.

’ INTRODUCTION

He I ultraviolet photoelectron spectroscopy (UPS) was firstdeveloped in the early 1960s,1,2 and it has been offering invalu-able insights into the electronic structures of molecules. Thephotoelectron kinetic energy distribution measured by UPSreveals the electronic energy level structure of a cation, whichis also related to the electron orbital energies of a neutralmolecule within Koopmans’ approximation.3 The vibrationalfine structure associated with each ionization band indicatesthe bonding character of the electron orbital from which anelectron has been removed; this permits a molecular orbital to beassigned to each photoelectron band.4 An ultraviolet photoelec-tron spectrometer with a high-performance hemispherical elec-tron energy analyzer and an intense He I lamp using the electroncyclotron resonance5 has achieved an impressively high spectralresolution (ΔE < 5 meV) that is limited only by the line width ofthe He I radiation. However, a hemispherical energy analyzer hasa small electron acceptance angle and a limited energy rangeobserved simultaneously; these factors severely limit the signalcount rate. Consequently, UPS has been scarcely applied tohighly diluted gases in supersonic molecular beams6 and biolo-gical molecules that could not be prepared in a large quantity.

Even at the highest spectral resolution achieved using ahemispherical energy analyzer,7,8 UPS fails to match the resolu-tion accomplished by pulsed-field-ionization zero-kinetic-energy

photoelectron spectroscopy (PFI-ZEKE).9,10 PFI-ZEKE ob-serves the resonances of the photon energy with the rovibronicenergies of Rydberg states associated with each cation state.11,12

It utilizes the fact that Rydberg states with high principalquantum numbers are structurally similar to a cationic statedue to the weak coupling between the Rydberg electron and theion core. PFI-ZEKE requires these Rydberg states to be long-lived; it thus encounters difficulties when the Rydberg statesrapidly dissociate or undergo autoionization.13 The higherexcited states of a cation, corresponding to electron removalfrom inner valence orbitals, tend to exhibit inherently broadphotoelectron bands, indicating the existence of rapid deactiva-tion processes. Therefore, PFI-ZEKE may not be the bestapproach for investigating these states.13 In this sense, He IUPS and PFI-ZEKE can be regarded as being complementary.

One advantage of UPS over PFI-ZEKE is that UPS can be usedto investigate the angular distribution of photoelectrons. Thephotoelectron angular distribution (PAD) is a useful tool in UPS.For example, Carlson and Anderson14 performed He I UPS ofbenzene in 1971 and found that its photoelectron bands exhibitcharacteristic angular anisotropy. This helped them to analyze

Received: October 14, 2010Revised: January 9, 2011

http://pubs.acs.org/action/showImage?doi=10.1021/jp1098574&iName=master.img-000.jpg&w=240&h=98

2954 dx.doi.org/10.1021/jp1098574 |J. Phys. Chem. A 2011, 115, 2953–2965

The Journal of Physical Chemistry A ARTICLE

the overlapping bands in the spectrum. The occurrence ofoverlapping photoelectron bands is an inherent characteristicof photoelectron spectroscopy of polyatomic molecules withmany valence electrons. As mentioned above, when the

vibrational structure of each photoelectron band can be resolved,it can be used for spectral assignment. However, when thestructure cannot be resolved (as is the case for inner valencephotoionization), PAD is valuable for characterizing the orbitalfrom which an electron is removed. The measurement of thephoton-energy dependence of PAD is even more useful forassignments.15

In this study, we report a He I photoelectron imaging (PEI)spectrometer16 that provides the ultimate collection efficiency ofphotoelectrons of 4π steradians and efficient measurement ofPADs. We use this novel spectrometer to perform UPS of thebenchmark systems of benzene and pyridine in supersonicmolecular beams. One disadvantage of PEI compared with ahemispherical energy analyzer is its relatively high susceptibilityto background photoemission from the instrument by scatteredHe I radiation. To overcome this difficulty, we designed a newelectrostatic lens that intercepts background photoelectronsemitted from an electrode of PEI toward an imaging detector.Compared with a hemispherical energy analyzer, a PEI spec-trometer achieves a considerably higher electron collectionefficiency at the expense of a lower energy resolution; how-ever, we demonstrate that an energy resolution of 0.735% isobtainable.

The PEI spectrometer developed in this study for He Iradiation is applicable to photoelectron spectroscopy with othervacuum ultraviolet (VUV) sources besides He I; such sourcesinclude synchrotron radiation, free electron lasers (FELs),17-19

and high harmonics generated by ultrashort pulsed lasers.20-22

In fact, PEI is considerably more difficult with incoherent He Ilight than coherent photon beams as the former cannot be tightlyfocused on a sample.

’EXPERIMENT

A.MolecularBeamChambers and ImagingSystem. Figure 1ashows the cross-sectional view of our experimental setup with thedimensions proportional to real sizes. We used a continuous orpulsed molecular beam depending on the experiment. In theformer case, a continuous supersonic jet was generated byexpanding a sample gas from a 25 μm diameter orifice at 294K. The jet was skimmed with a 0.8 mm diameter skimmer andintroduced into an ionization chamber. The source and ioniza-tion chambers were pumped by turbomolecular pumps withpumping rates of 2000 L/s (N2) and 820 L/s (N2), respectively.When Ar was continuously expanded at a stagnation pressure of1 atm, the ionization chamber pressure was ca. 7.3� 10-8 Torr.He I radiation at 21.22 eV (58.4 nm) was generated using acommercial discharge lamp (Omicron, HIS13) and it wastransmitted through a 0.8 mm diameter capillary. The end ofthe capillary was located ∼125 mm from the ionization point.The photon flux is 9.4� 1010 (photons/s)/mm2 at the samplingpoint. Electrons generated by photoionization of a sample wereaccelerated along the molecular beam propagation axis andprojected onto a 2D position-sensitive detector. The accelerationfield was created using stacked circular ring electrodes and thefield gradient was carefully adjusted to achieve a velocitymappingcondition.23 The details of the electrodes are described in thenext section. The entire regions of photoionization, acceleration,and flight of the electrons were shielded against external mag-netic fields by a permalloy tube. The position-sensitive detectorconsists of a chevron-type (dual) microchannel plate assembly(Hamamatsu, F1942-04; pore diameter, 25 μm; diameter,

Figure 1. (a) Cross-sectional view of our experimental setup. Thedimensions in this drawing are proportional to the real sizes. (b) Cross-sectional view (left half) and front view (right half) of our electrostaticlens system (all units in millimeters). A molecular beam is introducedfrom the bottom and is irradiated by He I radiation at the positionindicated by the cross (�). Electrodes 1-3 were spaced by a plasticinsulator, whereas electrodes 3-12 were spaced by 3 mm diameter rubyballs (indicated by red circles). The figure also shows an example of thevoltage settings for obtaining a photoelectron image of Kr.




77 mm) backed by a phosphor screen (P43) and a CCD camera(Andor, iXonEM DCL897; 512 � 512 pixels).In the time-gated experiments, 10-20% sample gases seeded

in He were expanded from a pulsed valve (General Valve; orificediameter: 100 μm) with a stagnation pressure of about 0.2-0.55atm. The front surface of the microchannel plate (MCP)assembly was maintained at the ground potential and the voltageapplied to the rear side was switched between 900 and 1400 Vusing a high voltage pulser (DEI, GRX-3.0K-H) to time-gate theMCP synchronously with a pulsed gas nozzle. The timing of thetrigger pulses for the gas nozzle, MCP, and CCD camera wascontrolled with a digital delay generator (Stanford ResearchSystems, DG535).B. Electrostatic Lens Design. Since He I radiation is inco-

herent with poor beam characteristics, its stray VUV light tendsto generate considerable background photoemission from metalsurfaces in the chamber, which interferes with VUV-PEI experi-ments. We estimated the angular divergence of the He I photonbeam to be(0.8�, which provides the 4.2 mm diameter of the HeI radiation at the ionization point and 5.4 mm at the exit side ofthe electron acceleration lens. We have tested various electrodesand found that the hole should be at least 5 times larger than theestimated VUV beam diameter. This indicates that there isconsiderable intensity of a halo around the main (incoherent)VUV beam. To avoid the light scattering of this halo, holes aslarge as 40� 35 mm rectangular size were made to electrode 4 inFigure 1b. While minimizing the scattering of VUV radiationdescribed above, we have searched the source of backgroundphotoemission. We experimentally tested various electrode de-signs and also simulated electron trajectories computationallywith the SIMION 3D software package (Scientific InstrumentServices). These investigations revealed that a major componentof the background photoemission originates from the repellerplate. Thus, we removed the solid repeller plate from our lenssystem.As shown in Figure 1b, our final design of the lens system

consists of a stack of 12 circular electrodes rigidly held bysupporting rods made of insulator. The electrodes numbered3-12 in the figure are spaced with 3 mm diameter ruby balls. Wereplaced one solid plate of a repeller with three components (nos.1-3 in Figure 1b). Electrode 1 is made from a high-transmission(90%) mesh (70 wires/inch) with a 6 mm diameter hole in thecenter for the molecular beam to propagate through. The meshreduces the cross-section of the electrode by 1 order of magni-tude compared with a solid plate, which suppresses backgroundphotoemission from this electrode. Furthermore, the negativevoltage of this electrode (no. 1) is set slightly lower in magnitudethan that of electrodes 2 and 3 (see Figure 1b). This retardationprevents photoelectrons produced by stray light of VUV radia-tion from being transmitted toward the detector, as shown inFigure 2.In Figure 1b, electrode 1 shields the acceleration fields against

the ground potential. Electrode 2 is held at the same voltage aselectrode 3 to avoid the field distortion caused by the largecentral hole in electrode 3. In the following, we refer to electrodes3-5 as the repeller, light port, and extractor, respectively.Voltages were independently applied to electrodes 1-5 using acomputer-controlled multiport power supply (MBS, A-1 Elec-tronics;(12 kVmax), whereas the other electrode voltages werepassively regulated by a register (22 MΩ) chain placed outsidethe vacuum chamber. The entire assembly can withstand voltagesup to 12 kV (we used up to 11 kV in this study).

The gradual change in the voltages across the electrodes 5-12results in “soft focusing” of the electron trajectories. Theprinciple is essentially the same as the velocity map imaging byEppink and Parker,23 while the strength of the acceleration fieldvaries much more gradually than the simplest design using threeelectrodes.23 Previously, Lin et al.24 have shown that soft focusingenlarges the source size of the charged particles from which theirvelocities can be focused on the detector. Lin et al. used an ionoptics consisting of 29 rings, and obtained 10mm o.d. (diameter)�0.3 mm (thickness) of the ion/electron source from which thevelocity can be mapped with the resolution (Δυ/υ) higher than1%. We examined computationally the soft focusing effect withdifferent numbers of lens electrodes in the acceleration region.Here we count the number of electrodes behind the extractor(lens no. 5): for instance, the electrodes shown in Figure 1b are

Figure 2. (a) Simulated trajectories (red, blue, and orange) of electronsemitted from the mesh electrode started from six different startingpositions spaced 10 mm apart with five different ejection angles spacedby 45�: for enhancing the visibility of the trajectories, the kinetic energyof each electron was set to 200 eV. The inset shows an expanded view.The black curves are the trajectories of photoelectrons from thephotoionization point, in which 500 electrons are ejected perpendicularto the flight axis ((Y axis direction) in theXY plane with 5.461 eV kineticenergy. The acceleration voltage on the repeller is -4000 V. Theequipotential lines are also shown (light brown). (b) 2D representationof the cylindrically symmetric potentialΦ(x,y) of (a). To show the turn-around trajectories of electrons emitted from the mesh electrode clearly,the trajectories of the electrons are stopped while they collided on themesh electrode. The enlarged view of electrodes nos. 1- 3 is in thefigure.




12 in total whereas the lens electrodes behind the extractor are 7(nos. 6-12). As shown in Figure 3, the calculated velocityresolution improves as more lens electrodes are employed;however, it was practically saturated at 7 electrodes. Therefore,we have decided to use 7 electrodes in our final design(Figure 1b). Ogi et al.25 have reported the electrode assemblywith 2 lens electrodes, which achieved a velocity resolution of0.1%. However, it was with a small ionization volume created by alaser (o.d. 0.1 mm� 2 mm). On the other hand, our system with7 lens electrodes described here achieves <0.37% for a consider-ably larger ionization volume.

’RESULTS AND DISCUSSION

A.He I PEI of RareGases. The overlap between themolecularbeam and the VUV light usually generates an approximatelycylindrical ionization volume. We assumed that the molecularand VUV light beams were respectively 3.4 and 4.2 mm indiameter and examined the electron velocity resolution (Δυ/υ)by SIMION 3D. When the molecular beam has a diametersimilar to or smaller than that of the VUV beam, the velocityresolutions parallel and perpendicular to the VUV beam arecomparable. If the molecular beam has a diameter significantlylarger than that of the VUV beam, the ionization volumebecomes asymmetric and the resolutions in the two directionsdiffer. The overall resolution of PEI is also limited by the imagingresolution, which is determined by the number of pixels of thecamera and real-time processing such as centroiding calculations.To fully exploit the performance of the charged particle optics,

the imaging resolution should be sufficiently high. Figure 4 showsthe photoelectron kinetic energy (PKE) distribution determinedfrom an inverse Abel-transformed image of He I photoionizationof Kr using a standard CCD camera (512� 512 pixels) withoutany resolution enhancement processing. The spin-orbit split-ting of Krþ (0.665 eV) is clearly resolved. The inset of Figure 4shows an expanded view of the PKE region 6.0-8.0 eV togetherwith a least-squares fit using a Gaussian function. The full widthat half-maximum (fwhm) of the Gaussian, 0.210 eV, corresponds

to an energy resolution (ΔE/E = 2Δv/v) of 2.9% at 7.22 eV. Thisresolution is insufficient to resolve the fine structure splitting of

Figure 4. PKE distributions determined by He I PEI of supersonicbeams of Kr using a 512 � 512 pixel CCD camera. The inset shows anexpanded view in the PKE region between 6.0 and 8.0 eV. The solid lineindicates the best-fit Gaussian to the observed data; it has a fwhm of 210meV at 7.22 eV.

Figure 5. (a) Velocity resolution evaluated by PEI of Kr as a function ofVext/Vrep. A low-resolution (512 � 512 pixels) CCD camera was usedwithout image processing. Due to the low resolution of the camera, thefocusing curve is broad. (b) Velocity resolution evaluated by PEI ofAr with a super-resolution (4096 � 4096 pixels) imaging system.The focusing curve is much sharper and the resolution profile for theVext/Vrep ratio is V-shaped. The error bars correspond to the fittingerrors (fwhm).

Figure 3. Calculated velocity resolution as a function of the number oflens electrodes behind the extractor. The ionization volume wasassumed to be 3.4 o.d. � 4.2 mm. For each number of electrodes, theapplied voltages were optimized to achieve the best resolution. Thegeometries of the electrodes are similar to Figure 1b except that the portelectrode (no. 4 in Figure 1b) was removed. This was because thepresence of the port electrode (no. 4) caused the voltage optimization tobe a more complex procedure. The distance between the repeller (no. 3)and extractor (no. 5) was maintained with the original design.

http://pubs.acs.org/action/showImage?doi=10.1021/jp1098574&iName=master.img-003.png&w=197&h=147





Ar (0.178 eV). Figure 5a shows the velocity resolution measuredas a function of Vext/Vrep by PEI of Kr using a standard CCDcamera. It shows that the resolution varies gradually with Vext/Vrep within the range of ca. 30 V about the best point. The broadfocusing curve and poor velocity resolution, which go against thetheoretical prediction, are attributed to the low imaging resolu-tion. To examine the focusing curve more precisely, we used asuper-resolution imaging system using a 2048� 2048 pixel CCDcamera.25 We achieved an effective imaging resolution of 4096�4096 by performing a 2 � 2 subpixel centroiding calculation.Figure 5b shows the velocity resolution measured by super-resolution imaging of Ar as a function of Vext (which was variedin 1 V steps) for Vrep and Vport fixed at -4000 and -3464.8 V,respectively. The response of the velocity resolution to thevariation in Vext is considerably sharper in this case; the resolu-tion clearly degrades on both sides of the best point. The bestvoltage ratio Vext/Vrep differs slightly for the two above-men-tioned measurements because the two measurements employedslightly different flight lengths. At the optimal ratios of Vport/Vrep = 0.8662 and Vext/Vrep = 0.8557, we obtained an energyresolution (ΔE/E) of 0.735% at 5.461 eV (fwhm: 40 meV). Thisvalue is in excellent agreement with the energy resolution(0.72%) calculated from 500 electron trajectories with initialvelocities parallel to the face of an MCP with 5.461 eV kineticenergy. The improved imaging resolution enabled the finestructure splitting of Ar to be resolved, as shown in Figure 6.In PEI experiments, the speed resolution Δv is limited by spatialresolution of the imaging system that is constant over theimaging area. Therefore, Δv/v varies inversely proportional tov. The energy resolution ΔE/E is 2 times larger than Δv/v bydefinition. Therefore, if the best energy resolution is 0.735% at5.461 eV, it degrades to 7.35% at 0.546 eV.With unpolarized light, the photoelectron angular distribution

is cylindrically symmetric about the light propagation direction.We analyzed the angular distribution [I(θ)] using the standardformula for single-photon ionization with unpolarized light:

IðθÞ ¼ σ

4π1þ β

232sin2 θ - 1

� ��

where θ is the angle between the light propagation directionand the outgoing photoelectron velocity, σ is the integrated

ionization cross section, and β is the anisotropy parameter. Wedetermined the anisotropy parameters for the 2P3/2 and

2P1/2ionization channels of Kr to be 1.24 ( 0.01 and 1.21 ( 0.01

Figure 6. PKE distributions determined by He I PEI of supersonicbeams of Ar using a super-resolution (4096 � 4096 pixels) imagingsystem. The inset shows an expanded view in the PKE region between5.1 and 5.6 eV. The solid line indicates the best-fit Gaussian to theobserved data; it has a fwhm of 40 meV at 5.461 eV.

Figure 7. (a) Photoelectron image obtained for He I photoionization ofa continuous supersonic beam of 8% benzene seeded in He (the back-ground image has been subtracted). The vertical band along the VUVlight path is due to photoionization of water vapor. (b) Ion imagemeasured in space-mapping mode to trace the spatial region wherethe He I light beam intersects the supersonic molecular beam. (c) Ionimage without the molecular beam. Comparison with (b) reveals thatthe bright elliptical spot in the center is due to benzene ions, whereas thevertical band along the propagation direction of the He I beam is due toresidual water in the ionization chamber. The arrow indicates thepropagation direction of the He I radiation.





respectively, from a symmetrized image of the original Cartesianimage. The errors are the fitting errors of the least-squares fitusing the above equation. We also determined the anisotropyparameters independently for the four quadrants in the photo-electron image and obtained 1.26 ( 0.04 and 1.22 ( 0.08 for2P3/2 and

2P1/2 respectively. The errors are the standard devia-tions of the anisotropy parameters determined from the fourquadrants; the larger errors obtained using this approach resultfrom the nonuniform sensitivity of the imaging detector. Theanisotropy parameters of Kr determined by both methods are ingood agreement with values given in the literature,26-28 whichindicates that symmetrization of the raw image prior to analysisdoes not affect the anisotropy parameter of our high-qualityimages and that it suppresses the effect of the nonuniformsensitivity of the detection system.

B. He I Photoelectron Imaging of Benzene and Pyridine.On the basis of results for rare gases, we performed PEIexperiments with polyatomic molecules in continuous super-sonic molecular beams. However, we found that the signal levelwas not sufficiently stronger than the background photoioniza-tion signal of the residual water vapor in the photoionizationchamber. The partial pressure of water in the ionization chamberwas evaluated to be ca. 2 � 10-8 Torr using a residual gasanalyzer (Stanford Research Systems, RGA200). Figure 7a showsa typical integrated photoelectron image obtained by He Iphotoionization of a continuous supersonic beam of 8% benzenein He with a stagnation pressure of 1.2 atm (the backgroundimage obtained without the molecular beam has been sub-tracted). The signal and background images had integrationtimes of 100 min. The image contains a vertical band along theVUV light path due to photoionization of water vapor. Thephotoelectrons from water are visible in Figure 7a, because imagesubtraction did not completely eliminate the background signaldue to the different conditions of the residual gas in the chamberwith and without the molecular beam. Parts b and c of Figure 7respectively show ion images observed with and without themolecular beam; mass selection was not performed. Comparisonof these two figures suggests that the bright elliptical spot in thecenter is due to benzene ions, whereas the vertical band along thepropagation direction of theHe I light originates fromwater ions.To enhance the contrast ratio between the signal and the

background, we used a pulsed beam instead of a continuousmolecular beam and we also increased the pumping rate byadding a turbomolecular pump (650 L/s) to the ionizationchamber. The He I radiation continuously induces photoioniza-tion of the residual gas. We time-gated both the CCD and theMCP so as to detect photoelectrons only when a pulsed beamwas introduced into the ionization region. A low-resolution CCDcamera (512� 512 pixels) was used for synchronization with anexternal trigger. As Figure 8a shows, pulsed operation consider-ably suppressed the background photoelectron from the residualwater vapor. Both the signal and background images wereintegrated for 90 min at a repetition rate of 20 Hz (the maximumread-out rate of our CCD camera). As the vacuum system allows100Hz of gas pulses at a stagnation pressure of 0.55 atm, the dutycycle of our PEI system can be easily increased by a factor of 5 bynot synchronizing the CCD camera.Figure 8b shows the photoelectron spectrum and the energy-

dependent anisotropy parameter measured for jet-cooled ben-zene. Benzene is the most fundamental aromatic molecule, andconsequently it has been used as a benchmark system for UPS ofpolyatomic molecules.14 As mentioned in the previous section,photoionization with unpolarized light generates a PAD that iscylindrically symmetric about the light propagation direction;therefore, a positive β corresponds to preferential ejection ofelectrons perpendicular to the symmetry axis, whereas a negativeβ corresponds to preferential ejection in the propagation direc-tion. No signature from benzene dimers (electron bindingenergy = 8.65 eV)29 is visible in our photoelectron spectrum.The modest resolution of 0.2-0.3 eV was too low to resolvevibrational structures. As Figure 8b shows, the energy-dependentanisotropy parameters determined in the present study are inexcellent agreement with values given in the literature;15,30,31

Table 1 tabulates and compares the numerical values withliterature values.15,30,31 In the first band 1e1g(π), the anisotropyparameter decreases at high binding energies. Carlson andAnderson14 have observed the same characteristic and suggested

Figure 8. (a) Symmetrized photoelectron image of He I photoioniza-tion of a pulsed supersonic beam of 18% benzene seeded in He (thebackground image has been subtracted). The left half is the raw imageand the right half is the slice image obtained by taking the inverse Abeltransform. With unpolarized light, the axis of cylindrical symmetry is thepropagation direction of the light (indicated by the arrow). The verticalband of the background signal observed with the continuous molecularbeam was suppressed. (b) Energy-dependent anisotropy parameters β(upper panel, solid circle) and photoelectron spectrum (lower panel)obtained from (a). Since the kinetic energy of the photoelectron isproportional to the square of the radius in the slice image, the energyresolution limited by the pixel size of a CCD camera degrades at higherenergies. Consequently, the energy widths considered for fitting β valueswere 60 and 140 meV in the region of binding energy around 19 and9 eV, respectively. Data from Sell et al.,30 Carlson et al.,15 and Mattssonet al.31 are given as open squares (0), open triangles (Δ), and openinverted triangles (r), respectively. The assignments of the ionizedorbitals are indicated at the vertical ionization point.15 The error bars arethe fitting errors.




Table 1. Electron Binding Energies (eV) and Anisotropy Parameters (β) for Benzene at 58.4 nm

this study Sell et al.c Carlson et al.d Mattsson et al.f

orbitala eBE β eBE β eBE β eBE β

1e1g(π) 9.11 1.11

9.21 1.21(3) 9.19 1.15 9.24 1.11(4)

9.29 1.08(6)b 9.33 1.04(4) 9.30 1.10 9.34 0.97(5)

9.38 1.06

9.44 1.14(5) 9.42 1.13(3) 9.48 1.05 9.44 0.90(4)

9.51 1.02(9) 9.53 1.01 9.53 0.84(9)

9.58 0.79(7) 9.60 0.85(7) 9.63 1.00 9.61 0.87(5)

9.71 0.83(7) 9.72 0.96

9.80 0.86(12) 9.84 0.97

3e2g(σ) 11.48 -0.23(2) 11.46 -0.13 11.49 -0.04(4)

11.53 -0.20(5) 11.59 -0.16 11.57 -0.01(2)

11.62 -0.18(3) 11.62 -0.04(7) 11.64 -0.07

11.75 -0.09(2) 11.70 -0.11(6) 11.73 -0.03 11.67 -0.02(3)

11.80 -0.04(5) 11.82 -0.04

11.88 -0.15(3) 11.90 -0.07(3) 11.90 0.02 11.84 0.06(2)

12.01 -0.17(3) 11.98 -0.05(5) 12.00 0.03 11.95 0.08(2)

12.08 -0.09(5) 12.05 0.09

1a2u(π) 12.07 0.20(3)

12.13 -0.16(3) 12.16 0.09(2) 12.16 0.18 12.19 0.23(2)

12.26 -0.07(2) 12.24 0.25(5) 12.25 0.18

12.33 0.23(3) 12.32 0.13 12.31 0.18(4)

12.39 0.00(3) 12.42 0.15(9) 12.38 0.11 12.44 0.19(4)

12.51 -0.04(4) 12.52 0.01(5) 12.49 0.15

12.63 -0.14(5) 12.60 0.14(5) 12.59 0.15 12.56 0.13(4)

12.69 0.08(5) 12.65 0.16 12.68 0.19(4)

12.77 0.00(7) 12.74 0.15

12.83 0.29

12.91 0.24

13.00 0.24

3e1u(σ) 13.82 0.11(4) 13.85 0.14(5) 13.89 0.20 13.85 0.34(5)

13.93 0.17(3) 13.94 0.21(4) 14.00 0.25 14.00 0.32(6)

14.04 0.20(2) 14.02 0.21(4) 14.06 0.22

14.15 0.14(2) 14.12 0.21(4) 14.17 0.20 14.14 0.28(5)

14.26 0.08(2) 14.20 0.14(7) 14.26 0.19 14.27 0.23(5)

14.30 0.11(2) 14.34 0.16

14.37 0.03(2) 14.38 0.07(5) 14.40 0.11 14.42 0.20(4)

14.48 -0.03(2) 14.46 -0.05(5) 14.49 0.02

1b2u(σ) 14.59 -0.01(2) 14.55 -0.04(5) 14.58 -0.07 14.56 0.11(3)

14.65 -0.20(4) 14.67 -0.20 14.68 -0.02(4)

14.70 -0.01(2) 14.73 -0.29(4) 14.76 -0.24

14.81 -0.05(2) 14.83 -0.26(3) 14.84 -0.20 14.81 -0.11(4)

14.91 -0.09(3) 14.91 -0.33(4) 14.94 -0.23 14.93 -0.10(4)

15.02 -0.08(4) 14.99 -0.25(8) 14.98 -0.13

15.09 -0.19(5) 15.07 -0.07 15.05 0.00(5)

15.12 0.02(4) 15.19 -0.08(13) 15.15 0.01 15.17 0.10(5)

15.22 0.06(4) 15.27 -0.06(9) 15.25 0.02

2b1u(σ) 15.32 0.11(4) 15.35 0.30(10) 15.34 0.07 15.28 0.27(6)

15.43 0.16(5) 15.46 0.34(8) 15.41 0.19 15.40 0.45(6)

15.53 0.18(5) 15.53 0.33(14) 15.50 0.33 15.52 0.43(5)

15.63 0.18(5) 15.62 0.32(7) 15.62 0.33 15.63 0.40(6)

15.72 0.15(6) 15.70 0.39(10) 15.74 0.25 15.74 0.42(5)

15.82 0.20(7) 15.81 0.14(7) 15.84 0.34 15.84 0.49(4)



that this variation may be due to the Jahn-Teller split states of D0.However, the effect of Jahn-Teller splitting on the photoelectronanisotropy parameters has not been elucidated experimentally ortheoretically.32 Another possible origin for the variation in βis a Coulomb phase. Since β for photoionization from 1e1g(π)increases with increasing PKE,15,33,34 β is expected to diminishat higher binding energies (i.e., lower PKEs) within the first band.The highest-resolution He I photoelectron spectrum of benzenereported by Baltzer et al.33 resolved the vibrational structure of1e1g(π) very clearly; unfortunately, however, β was not measuredwith a high vibrational resolution.Figure 9 shows the results for pyridine 10% seeded in He. The

energy-dependent photoelectron anisotropy parameters andphotoelectron spectrum were obtained by taking the inverse Abeltransform of the symmetrized image. The numerical values of β aretabulated and compared with values in the literature35,36 in Table 2.We found that a microdischarge was generated in theMCPwhen ahighdensity of neutralmolecules impinged on theMCP; to preventthis, we restricted the stagnation pressure to 0.2 atm. Micro-discharges have not been observed in other PEI spectrometers inour laboratory; they appear to be a specific problemof this detector.The low stagnation pressure of 0.2 atm necessitated using a long(3 h) integration time for pyridine for each signal and backgroundimage. Microdischarges could be completely eliminated by rede-signing the PEI spectrometer for amolecular beam traveling parallelto the face of the imaging detector.The first band between 9.2 and 10.2 eV corresponds to ioniza-

tion toD0 andD1 and the second band near 10.6 eV is assigned to

D2. D0 andD1 are due to electron removal from either 11a1(n) or1a2(π) orbitals; however, the ordering of these two states iscontroversial. ADC(3) Green function calculations37 predict a

Table 1. Continued

this study Sell et al.c Carlson et al.d Mattsson et al.f

orbitala eBE β eBE β eBE β eBE β

15.92 0.23(6) 15.90 0.32(8) 15.91 0.42

16.01 0.24(6) 15.98 0.38 (12) 16.02 0.52 15.96 0.49(4)

16.06 0.30(9) 16.07 0.52 16.07 0.46(5)

16.11 0.19(8) 16.16 0.50(9) 16.19 0.57(7)

16.20 0.09(9) 16.24 0.27(9) 16.30 0.49(8)

3a1g(σ) 16.84 0.38(5) 16.83 0.14(11) 16.84 0.57(6)

16.93 0.37(4) 16.91 0.45(4) 16.93 0.34 16.96 0.45(6)

17.01 0.26(5) 16.99 0.38(7) 17.03 0.40

17.09 0.43(8) 17.07 0.37 17.07 0.39(9)

17.19 0.21(4) 17.17 0.29 17.19 0.31(7)

17.26 0.21(4) 17.25 0.33

17.34 0.30(10) 17.34 0.38 17.31 0.23(11)

17.44 0.13(15) 17.42 0.21

17.52 0.30(8)

2e2g(σ) 18.89 -0.15(5)

18.95 -0.11(5)

19.01 -0.13(5)

19.07 -0.07(5)

19.13 -0.09(5)

19.19 -0.16(5) 19.20 -0.06(12)e

19.25 -0.15(5)

19.31 -0.09(6)

19.37 -0.03(6)aThe orbital assignments for benzene are those in ref 15. b Errors ((, in last digit unless indicated otherwise) given in parentheses. cThe energy-dependent anisotropy parameters are reproduced from ref 30. dThe energy-dependent anisotropy parameters are reproduced from ref 15.e β was determined by integrating over the area of the 2e2g (σ) band.

fReference 31.

Figure 9. Energy-dependent anisotropy parameter β (upper panel,solid circles) and photoelectron spectra (lower panel) obtained fromHe I photoionization of pulsed supersonic beams of 10% pyridineseeded in He. The image was symmetrized and the background imagewas subtracted prior to data analysis. Data from Piancastelli et al.36 andUtsunomiya et al.35 are given as open squares (0) and open triangles(Δ), respectively. Assignments of the ionized orbitals are indicated at thevertical ionization point.41 The error bars are the fitting errors.




Table 2. Electron Binding Energies (eV) and Anisotropy Parameters (β) for Pyridine at 58.4 nm

this study Utsunomiya et al.c Piancastelli et al.d

orbitala eBE β eBE β eBE β

11a1(n), 1a2(π) 9.26 0.19(6) 9.20 -0.08(5)

9.36 0.30(3) 9.30 0.17(5)

9.46 0.45(4) 9.40 0.07(5)

9.58 -0.06(10)b 9.56 0.52(3) 9.50 0.27(5)

9.66 0.56(1) 9.60 0.43(5)

9.72 0.48(3) 9.73 0.61(1) 9.70 0.42(5)

9.86 0.53(3) 9.80 0.65(2) 9.80 0.51(5)

9.90 0.58(2) 9.90 0.42(5)

10.00 0.40(5) 10.00 0.59(3) 10.00 0.47(5)

10.10 0.58(6) 10.10 0.42(5)

2b1(π) 10.42 0.49(6)

10.56 0.79(4) 10.54 0.91(3)

10.69 0.73(7) 10.64 0.77(2)

10.74 0.81(4)

7b2(σ) 12.39 -0.42(4) 12.37 -0.16(3)

12.51 -0.29(3) 12.48 -0.31(1)

12.63 -0.17(3) 12.65 -0.35(1)

12.75 -0.13(3) 12.75 -0.14(7)

12.88 -0.05(3)

1b1(π) 13.00 0.07(3) 13.07 0.12(4)

13.12 0.15(3) 13.17 0.26(3)

13.23 0.17(3) 13.27 0.27(6)

13.35 0.06(3) 13.37 0.27(2)

13.47 0.00(3) 13.47 0.37(3)

13.59 0.00(3)

10a1(σ) 13.70 0.01(3)

13.82 0.07(2)

13.93 0.06(2)

14.04 0.09(3)

14.15 0.11(3)

6b2(σ) 14.26 0.19(3)

14.37 0.29(2)

14.48 0.31(2)

14.59 0.32(2)

14.70 0.26(3)

14.81 0.23(4)

14.91 0.26(4)

15.02 0.25(4)

15.12 0.23(5)5b2(σ), 9a1(σ) 15.22 0.05(5)

15.32 -0.13(6)

15.43 -0.15(5)

15.53 -0.11(4)

15.63 -0.10(3)

15.72 -0.16(2)

15.82 -0.16(3)

15.92 -0.18(3)

16.01 -0.14(4)

16.11 -0.07(4)

16.20 -0.04(4)

16.30 -0.08(4)

16.39 -0.16(5)



π-n-π ordering, which is similar to the results of the Greenfunction38,39 and valence bond methods.40 However, morerecent calculations using SAC-CI,41 DFT-TP,42 CASPT2,43

and MRDCI44 suggest an n-π-π ordering. MRDCI predictedthat the first two ionic states have an energy difference of 0.7 eV,whereas the other methods have given differences only less than0.2 eV. In 1978, Utsunomiya et al.35 attempted to determine theordering of these two states using He I photoelectron spectros-copy; they compared the anisotropy parameters with those of2,6-lutidine (dimethylpyridine) for which spectroscopic assign-ments have been well established. Utsunomiya et al. observedthat the anisotropy parameter was ca. 0.2 near the ionizationthreshold, whereas it increased to ca. 0.6 in the middle of the firstband. On the basis of this, they assigned D0 of pyridine to the n

-1

state, since the n-1 state of 2,6-lutidine has a low β. Piancastelliet al.36 have performed similar measurements with synchrotronradiation but they varied the photon energy. They found that theanisotropy parameter increased when the photon energy wasincreased from 13 to 27 eV and that the higher binding energyregion of the first band exhibited a more rapid increase in β thanthe lower binding energy region. On the basis of this, they alsoassigned D0 to the n-1 state and D1 to the π-1 state.He I UPS creates PKEs of ca. 10 eV in ionization to D0 and D1.

In this PKE region, the photoelectron anisotropy parameterincreases slowly with increasing PKE due to the energy-depen-dent Coulomb phases.34 In contrast, the present study and theresults of Utsunomiya et al. reveal that the anisotropy parameterincreases within the first band with increasing binding energy(i.e., decreasing PKE). This suggests that the variation of β in thefirst band is not due to Coulomb phases but rather it is due to D1

being located close to D0; the contribution of D1 increases rapidlyin the first band causing β to increase. CMSXR calculations34

predict that β is almost zero at a PKE of ∼5 eV for bothionization to D0 andD1 but that it increases with increasing PKE;this explains the results of Piancastelli et al.36 The origin band of

D1 has not been determined yet. Appendix A shows a high-resolution He I spectrum of jet-cooled pyridine measured in ourlaboratory with a hemispherical energy analyzer.The fourth and fifth bands between 12.5 and 13.5 eV have

been assigned to 7b2σ and 1b1π, respectively.35 As far as we

know, anisotropy parameters for electron binding energies ofover 14 eV have not been reported. All these cation states are dueto the removal of an electron from the σ orbitals (see Figure 9).There are two different theoretical assignments for the over-lapping bands within 0.3 at 15.8 eV. These bands are associatedwith the 9a1(σ) C-H and 5b2(σ) C-C bonding orbitals. All thecalculations that predict n-π-π ordering for the first threecation states suggest 5b2(σ) < 9a1(σ), whereas the othermethods that support π-n-π ordering for the first three statessuggest the opposite result, 5b2(σ) > 9a1(σ). The assignments inFigure 9 follow the former assignment.C. Comparison of PEI with He I and SASE-FEL. We have

used the above-mentioned VUV-PEI spectrometer in photo-ionization experiments using a VUV free electron laser (SCSS:SPring-8 Compact SASE Source).45 As mentioned before, theresolutions of PEI using He I is limited by the ionization volume,whereas SCSS is free from this problem since the laser beam isfocused down to a diameter of ca. 0.1 mm. Our trajectorycalculations indicate that the energy resolution ΔE/E of ourcharged particle optics will be 0.04% at 5.461 eV for a smallionization volume created by a laser, which is close to theultimate resolution, 0.06%, obtainable using the best commer-cially available MCP that is 70 mm in diameter and has 10 μmpores. In practice, however, we found that the resolution issuperior with He I radiation than FEL radiation at 58.4 nm(Figure S1 in the Supporting Information), although the formerhas a much larger ionization volume. This is because SCSS, likeother FELs, uses self-amplified spontaneous emission (SASE) ofradiation that has unavoidable fluctuations in the spectrum,resulting in a large effective bandwidth of 0.1 eV.18

Table 2. Continuedthis study Utsunomiya et al.c Piancastelli et al.d

orbitala eBE β eBE β eBE β

16.48 -0.18(5)

16.57 -0.15(7)

8a1 (σ) 16.66 0.00(6)

16.75 0.14(6)

16.84 0.25(6)

16.93 0.32(5)

17.01 0.30(4)

17.10 0.35(4)

17.18 0.43(5)

17.27 0.44(6)

17.35 0.38(6)17.43 0.20(5)

7a1(σ) 19.54 0.07(7)

19.59 0.01(8)

19.64 0.00(8)

19.69 0.11(7)

19.75 0.15(8)aThe orbital assignments for pyridine are those made in ref 41. bErrors ((, in last digit unless indicated otherwise) given in parentheses. cReference 35.dThe energy-dependent anisotropy parameters are reproduced from ref 36.



VUV-FEL has the advantage of providing much more intenseradiation than He I. However, the pulse energy of the SCSSradiation was too high for single-photon ionization experiments;we had to use metal and gas filters to reduce it in our experiment.Since SCSS is operated at a low repetition rate (10-60 Hz), thisreduction in the pulse energy diminishes the advantage of VUV-FEL over He I for UPS.Since the VUV-FEL beam is tightly focused, we anticipated that

background photoelectrons have amuch smaller influence than forHe I. However, we found that the conventional three-plateelectrostatic lens23 generated considerable background electrons

with SCSS. Our new lens system was essential in obtaining high-quality electron images with SCSS.

’SUMMARY

WedescribedHe IUPS in the new formof PEI. The performanceof He I PEI was examined by observing photoionization of rare gasatoms and polyatomic molecules. We demonstrated that time-gatedmeasurement with a pulsed beam provides a high signal contrastrelative to the background from water vapor in the ionizationchamber. The photoelectron anisotropy parameters determinedfor benzene and pyridine were in good agreement with values givenin the literature. Real-time subpixel centroiding calculations achievedthe best energy resolution of 0.735% at 5.461 eV, which correspondsto a fwhmof 40meV.This does not represent the ultimate resolutionachievable with the spectrometer. It is possible to increase theresolution by 1 order of magnitude by tightly focusing the photon

Table A. Vibrational Assignments of the Bands in Figure 11

no. eBE (eV) exptl vibr energy (meV) assignment calcd vibr energy (meV) Δ (meV)a

1 9.199 0 0-0

2 9.273 74 1001 74 0

3 9.318 119 901 119 0

4 9.348 149 1002 148 -1

5 9.3925 193.5 901100

1 193 -0.5

6 9.419 220 1003 223 3

7 9.428 229 902 238 9

8 9.467 268 901100

2 267 -1

9 9.487 288 1004 294 6

10 9.504 305 902100

1 303 -2

11 9.541 342 901100

3 342 0

12 9.555 356 903 357 1

13 9.579 380 902100

2 379 -6

14 9.589 390 unknown

15 9.61 411 901100

4 416 5

16 9.626 427 903100

1 431 4

17 9.644 445 902100

3 454 9

18 9.662 463 not sure

19 9.697 498 903100

2 501 3

20 9.724 525 902100

4 537 12aΔ = (calcd vibr energy) - (exptl vibr energy).

Figure 10. He I photoelectron spectrum of jet-cooled pyridinemeasured with a hemispherical energy analyzer. The two bands withlowest electron binding energies are shown. He I radiation wasunpolarized and photoelectrons were detected in the perpendiculardirection with respect to the light propagation axis. The bandindicated by “11a1” is the origin band of the D0(n

-1)r S0 transition.The broad feature suggests that D1(π

-1) due to an electron removalfrom the 1a2 orbital exists within this band; however, its origin iscurrently unclear.

Figure 11. Expanded view of the He I photoelectron spectrum shownin Figure 10. The combination bands of the modes 9 and 10 (in-planetotally symmetric vibrations) are indicated.





beam. The CCD imaging detector in our instrument could bereplaced by a delay line detector; however, the resolution obtainableis comparable with that of a standard CCD camera. The highestimaging resolution is obtained by using a high-resolution camera andsuper-resolution image processing. However, it requires a high-frame-rate camera to avoid spatial overlapping between twoelectron impacts on the detector. In addition, either high-speedreal-time image processing to calculate the center of gravity foreach light spot or a large memory to store images for off-lineanalysis after the measurement are required. Our super-resolu-tion imaging system is currently able to handle up to 256 lightspots in a single frame (30 frames/s), which is too low for He IUPS. This problem will be solved when imaging devices withhigher frame rates and more rapid digital image processingbecome available.

’APPENDIX A: VIBRONIC STRUCTURE OF THE FIRSTBAND OF PYRIDINE

Innes et al.46 reviewed the electronic excited states of azaben-zenes and azanaphthalenes. They state that the adiabatic ioniza-tion energy of pyridine is 9.197 eV in the text but 74 740 cm-1

(9.267 eV) in their Table 7. The former value is from photo-electron spectroscopy of pyridine vapor by Reineck et al. (9.197eV) in 1982,47 whereas the latter value is probably from an earlierwork by Turner (9.26 eV) in 1970.48 Bolovinos et al. used 9.26eV in their analysis of Rydberg state spectroscopy,49 whereas laterstudies such as that by Walker et al.44 took the ionization energyof pyridine to be 9.197 eV. The difficulty in determining theionization energy is due to the small Franck-Condon factor ofthe 0-0 band, which is similar to the case of pyrazine that wepreviously studied.13

Figure 10 shows the He I photoelectron spectrum of jet-cooled pyridine measured in our laboratory with a hemisphericalenergy analyzer at a resolution of 10 meV. The overall features ofthis spectrum are in excellent agreement with those by Reinecket al.47 The D1 state is thought to be close to the D0 state; forinstance, SAC-CI calculations predict D0 and D1 to be 9.23 and9.36, respectively.41 This proximity is regarded as the cause of thespectral complexity of the first band; however, the origin of D1

has not been established.Figure 11 shows an expanded view of the first band. The

vibrational progressions of the mode 10 are clearly identified. Rieseet al. have performed mass-analyzed threshold ionization (MATI)spectroscopy of pyridine and deduced 74 185( 6 cm-1 (9.1978(0.0008 eV).50 MATI was performed within an excess vibrationalenergy of 0.2 eV from the origin.No examination has been reportedfor the binding energy region >9.4 eV where a broad backgroundappears in UPS. Table A lists the vibrational assignments for thespectrum. Notice that the first cold vibrational band 100

1 is at 9.273eV, which is regarded as the source of the old value (9.26 eV) of theionization energy. This illustrates the difficulty in experimentallydetermining the lowest cold band.

As for the Franck-Condon factor from S0, there is aninteresting difference for the 3s Rydberg state. Turner et al. haveperformed (2þ1) REMPI spectroscopy of pyridine via 3s Ryd-berg state in which they observed an intense band at 50 413 cm-1

(6.25 eV), and they assigned it to the 0-0 band.51 Dion andBernstein repeated the experiment in 1995,52 and they observedthe same band at 50 685 cm-1 (6.28 eV). Tsubouchi and Suzukihave performed (1þ20) REMPI photoelectron spectroscopyusing a femtosecond laser and observed the photoelectron signal

attributed to ionization via the 3s state.53 The photoelectronkinetic energy (PKE) they observed is related to the term value(T) of the Rydberg state as follows:

T ¼ PKEþ IE-hνprobe

where IE is the ionization energy and hνprobe is the photonenergy. Using an IE of 9.20 eV, T is calculated to be 6.27 eV.Thus, the assignment of the 3s state has been established. Noticethat the maximum of the Franck-Condon envelope in UPSappears at higher vibronic levels than the 0-0 band (Figures 10and 11), which contradicts the fact that two-photon excitation tothe 3s state exhibited an intensity maximum in the 0-0 band. Inthe case of pyrazine, we found that the energy gaps betweenD1(π

-1)-D0(n-1) and 3s(π-1)-3s(n-1) are different. It may

be that the proximity effect between the 3s(n-1) and 3s(π-1)states of pyridine differs from that between D0(n

-1) andD1(π

-1), which gives rise to different Franck-Condon factorsfor the 3s and D0 from S0.

’ASSOCIATED CONTENT

bS Supporting Information. Comparison of photoelectronkinetic energy distribution of Kr in photoionization by He Iradiation and FEL radiation and chromatic aberration of electro-static lenses. This material is available free of charge via theInternet at http://pubs.acs.org.

’AUTHOR INFORMATION

Corresponding Author*E-mail: [email protected].

Present Addresses^Department of Chemistry, Graduate School of Science andEngineering, Tokyo Metropolitan University, 1-1 Minami-Osawa,Hachioji, Tokyo 192-0397, Japan

’ACKNOWLEDGMENT

S.Y.L. thanks RIKEN for the Asia Program Associate scholar-ship to perform her Ph.D. at RIKEN. We thank Y. Ogi, T.Mizuno, and S. Kuroki for assistance with the experiments.

’REFERENCES

(1) Vilesov, F. I.; Kurbatov, B. L.; Terrenin, A. N. Soviet Phys.(Doklady) 1961, 6, 490.

(2) Turner, D. W.; AI-Jobory, M. I. J. Chem. Phys. 1962, 37, 3007.(3) Koopmans, T. C. Physica 1933, 1, 104.(4) Eland, J. H. D. Photoelectron Spectroscopy; Butterworth: London,

1974.(5) Baltzer, P.; Karlsson, L.; Lundqvist, M.; Wannberg, B. Rev. Sci.

Instrum. 1993, 64, 2179.(6) Pollard, J. E.; Trevor, D. J.; Lee, Y. T.; Shirley, D. A. Rev. Sci.

Instrum. 1981, 52, 1837.(7) €Ohrwall, G.; Baltzer, P.; Bozek, J. Phys. Rev. Lett. 1998, 81, 546.(8) €Ohrwall, G.; Baltzer, P.; Bozek, J. Phys. Rev. A. 1999, 59, 1903.(9) Jarvis, G. K.; Song, Y.; Ng, C. Y.; Grant, E. R. J. Chem. Phys. 1999,

111, 9568.(10) Merkt, F.; Osterwalder, A.; Seiler, R.; Signorell, R.; Palm, H.;

Schmutz, H.; Gunzinger, R. J. Phys. B, At. Mol. Opt. Phys. 1998, 31, 1705.(11) Schlag, E. W. Zeke Spectroscopy; Cambridge University Press:

Cambridge, 1998.(12) M€uller-Dethlefs, K.; Schlag, E. W.; Grant, E. R.; Wang, K.;

McKoy, B. V. Adv. Chem. Phys. 1995, 90, 1.



(13) Oku,M.;Hou, Y.; Xing, X.; Reed, B.; Xu,H.;Chang, C.;Ng, C. Y.;Nishizawa, K.; Ohshimo, K.; Suzuki, T. J. Phys. Chem. A 2008, 112, 2293.(14) Carlson, T. A.; Anderson, C. P. Chem. Phys. Lett. 1971, 10, 561.(15) Carlson, T. A.; Gerard, P.; Krause, M. O.; Grimm, F. A.; Pullen,

B. P. J. Chem. Phys. 1987, 86, 6918.(16) Suzuki, T. Annu. Rev. Phys. Chem. 2006, 57, 555.(17) Ayvazyan, V.; Baboi, N.; Bahr, J.; Balandin, V.; Beutner, B.;

Brandt, A.; Bohnet, I.; Bolzmann, A.; Brinkmann, R.; Brovko, O. I.;Carneiro, J. P.; Casalbuoni, S.; Castellano, M.; Castro, P.; Catani, L.;Chiadroni, E.; Choroba, S.; Cianchi, A.; Delsim-Hashemi, H.; Di Pirro,G.; Dohlus, M.; Dusterer, S.; Edwards, H. T.; Faatz, B.; Fateev, A. A.;Feldhaus, J.; Flottmann, K.; Frisch, J.; Frohlich, L.; Garvey, T.; Gensch,U.; Golubeva, N.; Grabosch, H. J.; Grigoryan, B.; Grimm, O.; Hahn, U.;Han, J. H.; Hartrott, M. V.; Honkavaara, K.; Huning, M.; Ischebeck, R.;Jaeschke, E.; Jablonka, M.; Kammering, R.; Katalev, V.; Keitel, B.;Khodyachykh, S.; Kim, Y.; Kocharyan, V.; Korfer, M.; Kollewe, M.;Kostin, D.; Kramer, D.; Krassilnikov, M.; Kube, G.; Lilje, L.; Limberg, T.;Lipka, D.; Lohl, F.; Luong, M.; Magne, C.; Menzel, J.; Michelato, P.;Miltchev, V.; Minty, M.; Moller, W. D.; Monaco, L.; Muller, W.; Nagl,M.; Napoly, O.; Nicolosi, P.; Nolle, D.; Nunez, T.; Oppelt, A.; Pagani,C.; Paparella, R.; Petersen, B.; Petrosyan, B.; Pfluger, J.; Piot, P.; Plonjes,E.; Poletto, L.; Proch, D.; Pugachov, D.; Rehlich, K.; Richter, D.;Riemann, S.; Ross, M.; Rossbach, J.; Sachwitz, M.; Saldin, E. L.; Sandner,W.; Schlarb, H.; Schmidt, B.; Schmitz, M.; Schmuser, P.; Schneider, J. R.;Schneidmiller, E. A.; Schreiber, H. J.; Schreiber, S.; Shabunov, A. V.;Sertore, D.; Setzer, S.; Simrock, S.; Sombrowski, E.; Staykov, L.; Steffen,B.; Stephan, F.; Stulle, F.; Sytchev, K. P.; Thom, H.; Tiedtke, K.; Tischer,M.; Treusch, R.; Trines, D.; Tsakov, I.; Vardanyan, A.; Wanzenberg, R.;Weiland, T.; Weise, H.; Wendt, M.; Will, I.; Winter, A.; Wittenburg, K.;Yurkov, M. V.; Zagorodnov, I.; Zambolin, P.; Zapfe, K. Eur. Phys. J. D2006, 37, 297.(18) Shintake, T.; Tanaka, H.; Hara, T.; Tanaka, T.; Togawa, K.;

Yabashi, M.; Otake, Y.; Asano, Y.; Bizen, T.; Fukui, T.; Goto, S.;Higashiya, A.; Hirono, T.; Hosoda, N.; Inagaki, T.; Inoue, S.; Ishii, M.;Kim, Y.; Kimura, H.; Kitamura, M.; Kobayashi, T.;Maesaka, H.;Masuda,T.; Matsui, S.; Matsushita, T.; Mar�echal, X.; Nagasono, M.; Ohashi, H.;Ohata, T.; Ohshima, T.; Onoe, K.; Shirasawa, K.; Takagi, T.; Takahashi.,S; Takeuchi, M.; Tamasaku, K.; Tanaka, R.; Tanaka, Y.; Tanikawa, T.;Togashi, T.; Wu, S.; Yamashita, A.; Yanagida, K.; Zhang, C.; Kitamura,H.; Ishikawa, T. Nat. Photonics 2008, 2, 555.(19) Barty, A.; Soufli, R.; McCarville, T.; Baker, S. L.; Pivovaroff,

M. J.; Stefan, P.; Bionta, R. Opt. Express 2009, 17, 15508.(20) Pfeifer, T.; Spielmann, C.; Gerber, G. Rep. Prog. Phys. 2006, 69,

443.(21) Krausz, F.; Ivanov, M. Rev. Mod. Phys. 2009, 81, 163.(22) Nisoli, M.; Sansone, G. Prog. Quantum Electron. 2009, 33, 17.(23) Eppink, A. T. J. B.; Parker, D. H. Rev. Sci. Instrum. 1997, 68,

3477.(24) Lin, J. J.; Zhou, J.; Shiu, W.; Liu, K. Rev. Sci. Instrum. 2003, 74,

2495.(25) Ogi, Y.; Kohguchi, H.; Niu, D.; Ohshimo, K.; Suzuki, T. J. Phys.

Chem. A 2009, 113, 14536.(26) Kreile, J.; Kurland, H. D.; Seibel, W.; Schweig, A. Nucl. Instrum.

Methods 1991, 308, 621.(27) Kreile, J.; Schweig, A. J. Electron Spectrosc. Relat. Phenom. 1980,

20, 191.(28) Karlsson, L.; Mattson, L.; Jadrny, R.; Siegbahn, K.; Thimm, K.

Phys. Lett. A 1976, 58, 381.(29) Krause, H.; Ernstberger, B.; Neusser, H. J. Chem. Phys. Lett.

1991, 184, 411.(30) Sell, J. A.; Kuppermann, A. Chem. Phys. 1978, 33, 367.(31) Mattsson, L.; Karlsson, L.; Jadrny, R.; Siegbahn, K. Phys. Scr.

1977, 16, 221.(32) Shiromaru, H.; Katsumata, S. Bull. Chem. Soc. Jpn. 1984, 57,

3543.(33) Baltzer, P.; Karlsson, L.; Wannberg, B.; €Ohrwall, G.; Holland,

D. M. P.; MacDonald, M. A.; Hayes, M. A.; von Niessen, W. Chem. Phys.1997, 224, 95.

(34) Suzuki, Y.; Suzuki, T. J. Phys. Chem. A 2008, 112, 402.(35) Utsunomiya, C.; Kobayashi, T.; Nagakura, S. Bull. Chem. Soc.

Jpn. 1978, 51, 3482.(36) Piancastelli, M. N.; Keller, P. R.; Taylor, J. W.; Grimm, F. A.;

Carlson, T. A. J. Am. Chem. Soc. 1983, 105, 4235.(37) Moghaddam, M. S.; Bawagan, A. D. O.; Tan, K. H.; von

Niessen, W. Chem. Phys. 1996, 207, 19.(38) von Niessen, W.; Diercksen, G. H. F.; Cederbaum, L. S. Chem.

Phys. 1975, 10, 345.(39) von Niessen, W.; Kraemer, W. P.; Diercksen, G. H. F. Chem.

Phys. 1979, 46, 113.(40) Tantardini, G. F.; Simonetta, M. Int. J. Quantum Chem. 1981,

20, 705.(41) Wan, J.; Hada, M.; Ehara, M.; Nakatsuji, H. J. Chem. Phys. 2001,

114, 5117.(42) Plashkevych, O.; Ågren, H.; Karlsson, L.; Pettersson, L. G. M.

J. Electron Spectrosc. Relat. Phenom. 2000, 106, 51.(43) Lorentzon, J.; F€ulscher, M. P.; Roos, B. O. Theor. Chim. Acta

1995, 92, 67.(44) Walker, I. C.; Palmer, M. H.; Hopkirk, A. Chem. Phys. 1990,

141, 365.(45) Liu, S. Y.; Ogi, Y.; Fuji, T.; Nishizawa, K.; Horio, T.;Mizuno, T.;

Kohguchi, H.; Nagasono, M.; Togashi, T.; Tono, K.; Yabashi, M.; Senba,Y.; Ohashi, H.; Kimura, H.; Ishikawa, T.; Suzuki, T. Phys. Rev. A 2010,81, 031403.

(46) Innes, K. K.; Ross, I. G.; Moomaw,W. R. J. Mol. Spectrosc. 1988,132, 492.

(47) Reineck, I.; Maripuu, R.; Veenhuizen, H.; Karlsson, L.; Sieg-bahn, K.; Powar, M. S.; Zu, W. N.; Rong, J. M.; Al-Shamma, S. H. J. Elec.Spectrosc. Relat. Phenom. 1982, 27, 15.

(48) Turner, D. W. Molecular Photoelectron Spectroscopy; Wiley:New York, 1970; p 323.

(49) Bolovinos, A.; Tsekeris, P.; Philis, J.; Pantos, E.; Andritsopoulos,G. J. Mol. Spectrosc. 1984, 103, 240.

(50) Riese, M.; Altug, Z.; Grotemeyer, J. Phys. Chem. Chem. Phys.2006, 8, 4441.

(51) Turner, R. E.; Vaida, V.; Molini, C. A.; Berg, J. O.; Parker, D. H.Chem. Phys. 1978, 28, 47.

(52) Dion, C. F.; Bernstein, E. R. J. Chem. Phys. 1995, 103, 4907.(53) Tsubouchi, M.; Suzuki, T. J. Phys. Chem. A 2003, 107, 10897.(54) Ure~na, F. P.; G�omez, M. F.; Gonz�alez, J. J. L.; Torres, E. M.

Spectrocim. Acta A 2003, 59, 2815.

Documents

He I Ultraviolet Photoelectron Spectroscopy of Benzene and ... · in Supersonic Molecular Beams Using Photoelectron Imaging ... pulsed molecular beam depending on the experiment